Published on

Published in: Health & Medicine, Technology
  • Be the first to comment

  • Be the first to like this

No Downloads
Total views
On SlideShare
From Embeds
Number of Embeds
Embeds 0
No embeds

No notes for slide


  1. 1. FINAL RESULTS OF THE EARLY TREATMENT FOR RETINOPATHY OFPREMATURITY (ETROP) RANDOMIZED TRIALBY William V. Good MD,* on behalf of the Early Treatment for Retinopathy of Prematurity Cooperative GroupABSTRACTPurpose: To present the final results of the Early Treatment for Retinopathy of Prematurity Study.Methods: Infants with bilateral high-risk prethreshold retinopathy of prematurity (ROP) (n = 317) had one eye random-ized to early retinal ablative treatment and the fellow eye managed conventionally (control eye). In asymmetric cases(n = 84), the eye with high-risk prethreshold ROP was randomized to early or to conventional management. High riskwas determined using a model based on the Cryotherapy for Retinopathy of Prematurity natural history cohort. Theprimary outcome was visual acuity assessed by masked testers using the Teller acuity card procedure. Structural exami-nations were performed at 6 and 9 months.Results: Grating acuity results showed a reduction in unfavorable visual acuity outcomes with earlier treatment, from19.8% to 14.3% (P < .005). Unfavorable structural outcomes were reduced from 15.6% to 9.0% (P < .001) at 9 months.Further analysis supported retinal ablative therapy for eyes with type I ROP, defined as zone I, any stage ROP with plusdisease; zone I, stage 3 ROP without plus disease; or zone II, stage 2 or 3 with plus disease. The analysis supported a“wait and watch” approach to type II ROP, defined as zone I, stage 1 and 2 without plus disease, or zone II, stage 3 with-out plus disease. These eyes should be considered for treatment only if they progress to type I ROP or threshold.Conclusion: Early treatment of high-risk prethreshold ROP significantly reduced unfavorable outcomes in both primaryand secondary (structural) measures.Trans Am Ophthalmol Soc 2004;102:233-250INTRODUCTION detachment.2 This degree of severity was termed the threshold for treatment of ROP, and was defined as atDespite the success of peripheral retinal ablation in least five contiguous or eight cumulative sectors (clockreducing the risk of retinal detachment, vision impair- hours) of stage 3 retinopathy of prematurity (ROP) inment still remains common in infants with severe zone I or II in the presence of plus disease (a degree ofretinopathy of prematurity (ROP).1 In the Cryotherapy for dilation and tortuosity of the posterior retinal bloodRetinopathy of Prematurity (CRYO-ROP) study, periph- vessels meeting or exceeding that of a standard photo-eral retinal ablation was administered when ocular find- graph).2 Treatment at threshold results in approximately aings indicated a risk of approximately 50% for retinal 50% reduction in the rate of retinal detachment. With the hope of improving this rate of unfavorableFrom the Smith-Kettlewell Eye Research Institute, San Francisco. A outcome, the timing indications for treatment of ROPcomplete list of the participants in the study appears in Appendix 2. The have been questioned, with some investigators advocatingauthors have no affiliation with or financial interest in the subject matter earlier treatment and others advocating conventionallyor materials discussed in the paper (eg, employment, consultancies, timed treatment.3,4 One concern with earlier treatment isstock ownership, honoraria), with the exception of Velma Dobson, PhD,who has received royalties from the sale of Teller acuity cards. the expected increase in surgical intervention in eyes withSupported by Cooperative Agreements (5U10 EY12471 and 5U10 ROP that would otherwise regress spontaneously. ThisEY12472) with the National Eye Institute of the National Institutes of concern has led to efforts to identify treatment selectionHealth, US Department of Health and Human Services, Bethesda, criteria that would result in earlier treatment only in thoseMaryland. eyes at highest risk for developing threshold ROP and/orPortions of this article were published in Archives of Ophthalmology an unfavorable visual or structural outcome in the absence(2003;121:1684-1694), copyright 2003, American Medical Association, of treatment.and are reprinted with permission. In 1999, the National Eye Institute funded a cooper-*Presenter. ative agreement to study early treatment for ROPBold type indicates AOS member. (ETROP).5,6 In the study, eyes of infants were randomizedTrans Am Ophthalmol Soc / Vol 102 / 2004 233
  2. 2. Good et alto early peripheral retinal ablation or standard treatment which meant that it was observed either until it reached(conventional management) if they developed prethresh- threshold and was treated or until the ROP regressedold ROP and if RM-ROP2, a risk analysis program based without progressing to threshold. In cases where only oneon natural history data from the CRYO-ROP study,7 indi- eye had reached high-risk prethreshold ROP, that eye wascated a high risk for an unfavorable outcome. randomized to treatment within 48 hours or to serve as aPrethreshold ROP was defined as any ROP in zone I that conventionally managed control, receiving treatment onlywas less than threshold; or in zone II stage 2 with plus if the ROP progressed to threshold severity. Infants indisease (dilation and tortuosity of posterior pole retinal whom either eye had developed threshold ROP prior tovessels in at least two quadrants, meeting or exceeding randomization were excluded from the study.that of a standard photograph2); or zone II, stage 3 disease For the analyses, eyes with prethreshold ROP thatwithout plus disease; or zone II, stage 3 with plus disease remained at low risk were categorized by the lowest zonebut fewer than five contiguous or eight cumulative clock and highest stage of ROP that ever developed. Eyes in thehours.2 This paper presents the final results of this study, randomized group were classified according to the zoneincluding implications for the timing of retinal ablative and stage of ROP that were present at the time oftreatment for ROP. Preliminary results were published randomization, as determined by the confirming exam-previously.8 iner’s observations.METHODS Functional Outcome The functional outcome of each randomized eye at 9Study protocols were approved by the review boards of all months corrected age was evaluated by assessment ofparticipating institutions, and parents provided written monocular grating acuity, conducted by one of two testersinformed consent prior to infants’ enrollment into the masked to the eye’s treatment assignment, who traveled tostudy and again at randomization. Details of the study the study centers for and laser technique are described elsewhere.6 The technique employed to evaluate grating acuity Infants with birth weights less than 1,251 g and birth was the Teller acuity card procedure9,10 as used previouslydates between October 1, 2000, and September 30, 2002, in the CRYO-ROP study.11,12 Acuity was scored as thewere screened at 26 participating centers. If an infant spatial frequency of the finest grating to which the infantdeveloped ROP, parents were asked to consent to data showed a consistent fixation response. Eyes in whichcollection and possibly increased frequency of examina- visual acuity was too poor to be quantified in this way weretions. Study-certified ophthalmologists conducted serial categorized as having no light perception (NLP), lightexaminations to detect rate of progression of ROP, devel- perception only (LP), or detection of the grating on theopment of prethreshold ROP, and development of thresh- low vision (LV) card only. The LV card has 2.2-cm-wideold ROP. If at least one eye reached prethreshold ROP, black-and-white stripes covering one half of the card. Itthe infant’s demographic and ROP information was was not used to quantify vision, but only to determineentered into the RM-ROP2 risk model to determine the whether the infant had pattern vision. The tester waslikelihood of progression to an unfavorable outcome in the permitted to move the LV card and/or to present it at anyabsence of treatment.7 distance and at any location in the infant’s visual field. The risk determination was made at the coordinating Visual acuity data were included in analyses only ifcenter, using the RM-ROP2 model7 to evaluate data the following criteria were met: (1) an acuity result (meas-provided by the clinical center. If the risk of progression urable acuity, detection of the grating on the LV card, LP,to an unfavorable outcome in the absence of treatment or NLP) was obtained for each eye of bilateral high-riskwas calculated to be ≥15%, consent for the randomized prethreshold cases or the randomized eye of asymmetrictrial was obtained, and randomization occurred. These cases; (2) treatment for amblyopia, if present, had beeneyes that had a risk of ≥15% were termed “high-risk” prescribed for at least 4 weeks prior to the acuity test; andprethreshold. Eyes with <15% risk were termed “low- (3) refractive error, if present in either eye of bilateralrisk” and were followed every 2 to 4 days for at least 2 high-risk prethreshold cases or in the randomized eye inweeks until the ROP regressed, or the risk progressed to asymmetric cases, had been corrected for at least 2 weeks≥15%. If both eyes were eligible for randomization, one prior to the acuity test. The criteria for correction ofeye was assigned at random to earlier treatment with abla- refractive errors were myopia greater than –4.00 diopterstive therapy within 48 hours of the first diagnosis of high- (D), hyperopia greater than +5.00 D, and/or astigmatismrisk prethreshold ROP. Treatment was generally laser greater than 2.50 D in one or both eyes. Correction oftherapy, but cryotherapy was allowed. The fellow eye anisometropia greater than 1.50 D spherical equivalent orserved as the control and was managed conventionally, 1.50 D cylinder was required only if the examining physi-234
  3. 3. Final Results of the Early Treatment for Retinopathy of Prematurity (ETROP) Randomized Trialcian found evidence of amblyopia. based sample size on structural outcome. This was a The visual acuity outcome was divided into four cate- conservative approach, since in the CRYO-ROP study,gories of functional response: normal, defined as greater unfavorable functional outcome rates were approximatelythan or equal to 3.70 cycles per degree13,14; below normal, 50% higher than unfavorable structural outcome rates atdefined as 1.85 to less than 3.70 cycles per degree (from ages at which functional outcome was tested.1,17,18approximately 4 to ≥2 standard deviations below the mean The statistical technique used to compare the eyesgrating acuity for a 9-month-old child14); poor, if less than treated at high-risk prethreshold with the conventionally1.85 cycles per degree but measurable with one of the managed high-risk prethreshold eyes was developed andstandard acuity cards (not the LV card); and blind/low used in the CRYO-ROP study.19 It combines the data fromvision (NLP, LP only, or LV only). These functional infants with bilateral disease (both eyes eligible) andoutcome categories of grating acuity results were further asymmetric disease (one eye eligible) into one overall chi-grouped into “favorable” and “unfavorable” designations. square analysis of outcome differences between the twoThe favorable grouping included eyes in the normal and treatment groups. Although not part of the original studybelow normal categories. The unfavorable grouping design, functional and structural results are also presentedincluded eyes in the poor and blind/low vision categories, by the International Classification of ROP (ICROP) andwhich would be expected to have a poor long-term prog- by RM-ROP2 categories to allow a more detailed exami-nosis for visual function.15 nation of the data. A Data and Safety Monitoring Committee ofStructural Outcome researchers, clinicians, and an ethicist not directlyStructural outcome was documented with a dilated fundus involved in the ETROP study met in person every 6examination at 6 months and 9 months corrected age by months to review adverse event and outcome data and tostudy-certified examiners. Complete ophthalmologic exam- monitor study progress. The committee approved theinations were performed at both of these ages; at the 9- protocol and monitored the performance of participatingmonth examination, a developmental questionnaire centers.(DDST16; results not reported in this article) was conducted.Refractive errors were determined by cycloplegic RESULTSretinoscopy after instilling 1% cyclopentolate hydrochloride.When there was a medical contraindication to this drop, At the 26 clinical sites, 828 infants whose parents hadeither 0.5% cyclopentolate or 1% tropicamide was used. At given consent for systematic follow-up of ROP were iden-6 months, an unfavorable outcome was defined as (1) a tified as having prethreshold disease in one or both eyes.posterior retinal fold involving the macula, (2) a retinal Among the 828 infants with prethreshold ROP, there weredetachment involving the macula, or (3) retrolental tissue or 499 (60%) whose eye or eyes were classified as high-risk“mass” obscuring the view of the posterior pole. If an infant and who were thereby eligible for the randomized trialrequired a vitrectomy or scleral buckle, the 6-month exami- (Figure 1). Among these 499 infants, consent for random-nation was conducted prior to the surgery. At the 9-month ization was not obtained for 40 infants, and high-riskexamination, eyes that had received a vitrectomy or scleral prethreshold ROP was not confirmed by the requiredbuckle were classified for study purposes as having an unfa- second study-certified examiner or for other reasons in 58vorable structural outcome.Statistical AnalysesThe ETROP study was designed to detect a 35% reduc-tion in the percentage of eyes having an unfavorable struc-tural outcome with a type I error rate of 0.05 and a powerof 80%.6 Using data from the CRYO-ROP study, thepercentage of unfavorable eyes managed conventionallywas predicted to be 20%.6 If earlier treatment produced a35% reduction, 13% of the earlier treated eyes wouldhave an unfavorable outcome. Taking into account thatapproximately 80% of infants were expected to have botheyes eligible for the study, the number of infants neededfor the study was 370.6 The primary outcome for this study FIGURE 1was visual function, for which there are limited data on Flow chart indicating status of the 828 infants in the ETROP study whowhich to conduct sample size calculations. Therefore, we developed prethreshold retinopathy of prematurity in one or both eyes. 235
  4. 4. Good et al TABLE 1. RISK STATUS BY ICROP CATEGORY FOR THE EYE WITH HIGHER RISK OF POOR STRUCTURAL OUTCOME FOR ALL 828 INFANTS WITH PRETHRESHOLD RETINOPATHY OF PREMATURITY IN ONE OR BOTH EYES* AT LEAST ONE EYE RISK IN PRETHRESHOLD EYE ICROP CATEGORY PRETHRESHOLD LOW RISK HIGH RISKZONE STAGE PLUS DISEASE NO. OF PATIENTS NO. OF PATIENTS NO. %I 3 Yes 29 0 29 100I 3 No 23 1 22 95.7I 1 or 2 Yes 19 0 19 100I 1 or 2 No 117 9 108 92.3II 3 Yes 271 13 258 95.2II 3 No 303 295 8 2.6II 2 Yes 66 11 55 83.3Total 828 329 499 60.3ICROP, International Classification of Retinopathy of Prematurity.20*Risk of poor structural outcome was based on the RM-ROP2 risk model analysis for each eye (a risk analysis program based on natural history data from the Multicenter Trial of Cryotherapy for Retinopathy of Prematurity Study).7 Low risk was less than 0.15; high risk was 0.15 or greater. Plus disease was defined as a degree of dilation and tortuosity of the posterior retinal blood vessels meeting or exceeding that of a standard photograph.2infants. Thus, 401 infants were enrolled into the random- 3, without plus disease were high-risk. The parallelsized trial. The remaining 329 infants, who had prethresh- between the RM-ROP2 model and ICROP are particu-old ROP that was judged to be low-risk, were followed as larly striking even though the former takes into account aclinically indicated and then underwent follow-up study number of demographic and disease-related factors thatexaminations at 6 months corrected age to determine reti- are not part of ICROP.nal outcomes. Table 2 provides baseline characteristics for the 401 Table 1 shows the distribution of prethreshold eyes by infants who entered the randomized trial. The mean birthRM-ROP2 risk classification7 and by severity of prethresh- weight was 703 g and the mean gestational age was 25.3old ROP according to the ICROP20 characteristics. One weeks. At the time of randomization, 79.1% of the infantseye per infant is represented in Table 1, and that is the eye had RM-ROP2 high-risk prethreshold disease bilaterally.with the higher risk, according to the RM-ROP2 model.7 The remaining 20.9% of infants had asymmetric disease,Table 1 shows that the ICROP can serve as a good indica- with high-risk prethreshold ROP in only one eye; thetor of most of the high-risk prethreshold eyes. For fellow eye had less severe ROP.prethreshold zone I eyes with plus disease, 100% were Table 3 shows the distribution of eyes treated at high-high-risk; when stage 3 was present without plus disease, risk prethreshold and conventionally managed (control)95.7% were high-risk; and with stage 1 or 2, without plus eyes by ICROP categories at randomization, along withdisease, 92.3% were high-risk. In zone II, 95.2% of eyes the percentage of conventionally managed eyes thatthat were classified as stage 3 with plus disease and 83.3% reached threshold ROP. Zone I disease accounted forof eyes with stage 2 with plus disease were high-risk, approximately 40% of randomized eyes. The largest cate-whereas only 2.6% of the 303 eyes that were zone II, stage gories of high-risk prethreshold eyes were those with zone II, stage 3, with plus disease (42.1% of prethresholdTABLE 2. BASELINE CHARACTERISTICS OF 401 RANDOMIZED PATIENTS treated eyes and 43.7% of conventionally managed eyes), and those with zone I, stage 1 or 2 with no plus diseasePatients with bilateral high-risk prethreshold ROP 79.1 (27.4% of prethreshold treated eyes and 26.1% andBirth weight, mean ± SD, g 703 ± 148Gestational age, mean ± SD, wk 25.3 ± 1.4 conventionally managed eyes). Table 3 also indicates thatMale 54.4 66.4% of eyes in the conventionally managed groupSingleton births 71.1 progressed to threshold and underwent peripheral retinalBorn in the study hospital 80.3 ablation at that time.Race White 63.8 The average age at high-risk prethreshold treatment African American 18.0 was 35.2 weeks postmenstrual age (SD, 2.3; range, 30.6 to Hispanic 14.7 42.1 weeks) and 10.0 weeks chronological age (SD, 2.0). Other 3.5 The average age for treatment of eyes in the convention-ROP, retinopathy of prematurity. ally managed group that went on to threshold was 37.0*Data are presented as percentage unless otherwise indicated. weeks postmenstrual age (SD, 2.5; range, 31.9 to 46.6 weeks) and 11.9 weeks chronological age (SD, 2.2). Only236
  5. 5. Final Results of the Early Treatment for Retinopathy of Prematurity (ETROP) Randomized Trial TABLE 3. DISTRIBUTION OF SEVERITY OF ROP AT RANDOMIZATION IN EYES WITH HIGH-RISK PRETHRESHOLD ROP AND PERCENTAGE OF CONVENTIONALLY MANAGED EYES WITH HIGH-RISK PRETHRESHOLD ROP THAT REACHED THRESHOLD* EYES TREATED AT HIGH-RISK CONVENTIONALLY ICROP CATEGORY AT PRETHRESHOLD MANAGED EYES RANDOMIZATION (n = 361) (n = 357) REACHINGZONE STAGE PLUS DISEASE NO. (%) NO. (%) THRESHOLD (%)I 3 Yes 19 (5.3) 19 (5.3) 94.7I 3 No 17 (4.7) 16 (4.5) 68.8I 1 or 2 Yes 14 (3.9) 16 (4.5) 87.5I 1 or 2 No 99 (27.4) 93 (26.1) 49.5II 3 Yes 152 (42.1) 156 (43.7) 71.8II 3 No 5 (1.4) 4 (1.1) 25II 2 Yes 55 (15.2) 53 (14.8) 66Total 361 (100) 357 (100) 66.4ICROP, International Classification of Retinopathy of Prematurity20; ROP, retinopathy of prematurity.*High risk was 0.15 or greater. Plus disease was defined as a degree of dilation and tortuosity of the posterior retinal blood vessels meeting or exceed- ing that of a standard eye received cryotherapy at threshold ROP as the (P < .005). Within-subject comparison afforded a power-primary treatment for ROP, with the others receiving laser ful opportunity to examine treatment effects whileretinal ablation. Some eyes received supplementary controlling for individual characteristics. Results from thecryotherapy at the time of the initial treatment. 33 infants with bilateral disease in whom there wereRetreatment was conducted in 13.9% of eyes treated at discordant outcomes in the two eyes provide evenhigh-risk prethreshold ROP and in 11.0% of convention- stronger evidence of a beneficial effect of treatment atally managed eyes treated at threshold ROP. high-risk prethreshold ROP (P < .005). Thirty-seven infants with bilateral high-risk prethreshold disease hadPrimary Outcome an unfavorable outcome in both eyes.In this final report, grating visual acuity data have been Table 5 provides a more detailed presentation of theobtained from 372 infants (98.2% of patients that grating acuity results. Although differences were notsurvived). Data were not obtained from 22 infants who statistically significant in these smaller categories, theredied prior to the 9-month examination and seven infants were more high-risk prethreshold treated eyes thanwhose parents did not bring them for the examination. conventionally managed eyes that had grating acuity inAverage corrected age (age from expected date of deliv- the normal range for age (P = .38). In addition, fewer eyesery) at the time of grating acuity assessment was 10.3 randomized to high-risk prethreshold treatment thanmonths (SD, 1.8). conventionally managed eyes were designated as blind or Table 4 presents the proportion of randomized eyes low vision (P = .07).with unfavorable grating acuity outcomes at 9 months.Overall, there was a significant benefit of treatment of Secondary Outcomeseyes at high-risk prethreshold ROP, with a reduction in Structural outcome data have been obtained from 366unfavorable visual acuity outcome from 19.8% to 14.3% infants (94.8% of patients that survived) at 6 months TABLE 4. NINE-MONTH GRATING ACUITY OUTCOME FOR RANDOMIZED PATIENTS* EYES TREATED AT HIGH-RISKEYES TREATED PRETHRESHOLD CONVENTIONALLY MANAGED EYES χ2 P VALUEBilateral 292 (15.4) 292 (21.2) 8.76† <.005Asymmetric 44 (6.8) 36 (8.3) 0.07 .80Total 336 (14.3) 328 (19.8) 8.03 <.005*Data are presented as number (percentage unfavorable) unless otherwise indicated.†Based on discordant pairs (25 infants with favorable outcomes in earlier treated eyes and unfavorable outcomes in conventionally managed eyes; eight infants with unfavorable outcomes in earlier treated eyes and favorable outcomes in conventionally managed eyes). 237
  6. 6. Good et al TABLE 5. DISTRIBUTION OF NINE-MONTH GRATING ACUITY OUTCOMES AMONG RANDOMIZED EYES BY TREATMENT ASSIGNMENT* EYES TREATED AT HIGH-RISK PRETHRESHOLD CONVENTIONALLY MANAGED EYESOUTCOME (n = 336) (n = 328)Favorable Normal (≥3.70 cycles per degree) 219 (65.2) 203 (61.9) Below normal (1.85 to <3.70 cycles 69 (20.5) 60 (18.3) per degree)Unfavorable Poor (measurable but <1.85 cycles 15 (4.5) 18 (5.5) per degree) Blind/LV (NLP, LP only, LV card only) 33 (9.8) 47 (14.3)Total 336 (100) 328 (100)LP, light perception; LV, low vision; NLP, no light perception.*Data are presented as number (percentage).corrected age and 372 infants (98.2% of patients that case for grating acuity outcome, results from infants withsurvived) at 9 months corrected age. Six-month data were bilateral disease in whom there were discordant outcomesnot obtained from 15 infants who died prior to the exam- in the two eyes provide strong evidence of a beneficialination and 20 infants whose parents did not bring them effect of treatment at high-risk for the examination. Average corrected age at the 6- Among the 30 high-risk prethreshold treated eyesmonth examination was 5.5 months (SD, 2.2). At 9 that had an unfavorable structural outcome at 9 months,months, structural outcome data were not obtained from two had a partial retinal detachment involving the macula,22 infants who died prior to the examination and seven 23 had undergone a vitrectomy or a scleral buckle, andinfants whose parents did not bring them in for the exam- five had total retinal detachment. Among the 51 conven-ination. Average corrected age at the 9-month examina- tionally managed eyes with an unfavorable outcome, fourtion was 9.8 months (SD, 1.4). This is younger than the had a partial retinal detachment involving the macula, 43average age for acuity testing because the final acuity data had undergone a vitrectomy or a scleral buckle, and fourwere sometimes collected during retesting after the 9- had total retinal detachment.month structural examination. Structural outcome results at the 6-month examina- The results for the 9-month structural outcome are tion for eyes randomized at high-risk prethreshold ROPpresented in Table 6. Data indicate a statistically signifi- were similar to those at 9 months, as shown in Table 7. Six-cant benefit of treatment of eyes with high-risk prethresh- month structural outcome data were also collected forold ROP, with unfavorable structural findings reduced low-risk prethreshold eyes (determined by the RM-ROP2from 15.6% in conventionally managed eyes to 9.0% in program to have a <15% risk for an unfavorable outcome).high-risk prethreshold treated eyes (P < .001). As is the Among this group of 329 infants, 51 (15.5%) had at least TABLE 6. NINE-MONTH STRUCTURAL OUTCOME FOR RANDOMIZED PATIENTS* EYES TREATED AT CONVENTIONALLYEYES TREATED HIGH-RISK PRETHRESHOLD MANAGED EYES χ2 P VALUEBilateral 290† (10.3) 291† (17.2) 11.7‡ <.001Asymmetric 42§ (0) 36 (2.8) 1.2 .28Total 332 (9.0) 327 (15.6) 12.6 <.001¶*Data are presented as number (percentage unfavorable) unless otherwise indicated.†Less than 293 because of inability to determine the structural outcome.‡Based on discordant pairs (25 infants with favorable outcomes in earlier treated eyes and unfavorable outcomes in conventionally managed eyes; 6 infants with unfavorable outcomes in earlier treated eyes and favorable outcomes in conventionally managed eyes).§Less than 43 because of inability to determine the structural outcome.¶Twenty-four eyes with partial retinal detachment not including the macula (stage 4A) had vitrectomy or a scleral buckling procedure prior to the 9-month examination and are included in this Table as having an unfavorable outcome. When the analysis was based on the structural outcome of these examina- tions, P = .002. Stage 4B or 5 eyes were a priori considered unfavorable in this study.238
  7. 7. Final Results of the Early Treatment for Retinopathy of Prematurity (ETROP) Randomized Trial TABLE 7. SIX-MONTH STRUCTURAL OUTCOME FOR RANDOMIZED PATIENTS* EYES TREATED AT CONVENTIONALLYEYES TREATED HIGH-RISK PRETHRESHOLD MANAGED EYES χ2 P VALUEBilateral 286† (5.6) 283† (11.0) 8.3‡ .004Asymmetric 39§ (0) 36 (2.8) 1.1 .29Total 325 (4.9) 319 (10) 9.2 .002*Data are presented as number (percentage unfavorable) unless otherwise indicated.†Less than 290 because of inability to determine the structural outcome.‡Based on discordant pairs (21 infants with favorable outcomes in earlier treated eyes and unfavorable outcomes in conventionally managed eyes; 6 infants with unfavorable outcomes in earlier treated eyes and favorable outcomes in conventionally managed eyes).§Less than 40 because of inability to determine the structural outcome. TABLE 8. GRATING ACUITY AT NINE MONTHS FOR INFANTS WITH BILATERAL HIGH-RISK PRETHRESHOLD RETINOPATHY OF PREMATURITY BY ICROP CATEGORY AND RM-ROP2 RISK* ICROP CLASSIFICATION EYES TREATED AT CONVENTIONALLY DISCORDANT PAIRSZONE STAGE PLUS DISEASE HIGH-RISK PRETHRESHOLD MANAGED EYES A† B‡I 3 Yes or No 26 (30.8) 26 (53.8) 7 1I 1 or 2 Yes 9 (22.2) 9 (22.2) 0 0I 1 or 2 No 76 (10.5) 76 (15.8) 4 0II 3 Yes 111 (15.3) 111 (18.0) 9 6II 3 No 3 (0) 3 (0) 0 0II 2 Yes 34 (14.7) 34 (17.6) 2 1RM-ROP2 risk0.15 to <0.30 110 (11.8) 110 (12.7) 4 30.30 to <0.45 75 (13.3) 75 (20.0) 7 2≥0.45 81 (22.2) 81 (34.6) 13 3ICROP, International Classification of Retinopathy of Prematurity20; RM-ROP2,7 a risk analysis program based on natural history data from theMulticenter Trial of Cryotherapy for Retinopathy of Prematurity Study.*Data are presented as number (percentage unfavorable) unless otherwise indicated. High risk was 0.15 or greater. Plus disease was defined as a degree of dilation and tortuosity of the posterior retinal blood vessels meeting or exceeding that of a standard photograph.2†For group A, earlier treated eyes had a favorable outcome, and conventionally managed eyes had an unfavorable outcome.‡For group B, earlier treated eyes had an unfavorable outcome, and conventionally managed eyes had a favorable eye that progressed to the conventional threshold for disease.treatment and was treated accordingly. An unfavorable As shown at the bottom of Tables 8 and 9, examina-outcome occurred in only 1.3% of the 302 low-risk tion of outcome by RM-ROP2 risk category showedprethreshold eyes for which 6-month structural outcome greater benefit in both grating acuity and structuraldata were available. outcomes for earlier treatment in high-risk prethreshold eyes with ≥30% risk than in high-risk prethreshold eyesRelationship to ICROP Classification with 15% to <30% risk.Tables 8 and 9 present the visual acuity and structuraloutcomes for randomized eyes, stratified by ICROP cate- Other Ocular and Clinical Findingsgory and by RM-ROP2 risk category. The greatest benefit The distribution of refractive errors at the 9-month exam-of treatment at high-risk prethreshold ROP versus ination was similar between the high-risk prethresholdconventional management occurred in eyes that had zone treated eyes that received early treatment and those thatI, stage 3 ROP, with and without plus (30.8% unfavorable were conventionally managed.versus 53.8% unfavorable). A relative benefit from inter- Cataract/aphakia that was not associated with totalvention at high-risk prethreshold ROP for both visual retinal detachment or vitrectomy was found in four eyesacuity and structural outcomes was also seen among eyes (1.2%) in the group treated at high-risk prethreshold andthat had zone I, stage 1 or 2 ROP without plus disease, in four eyes (1.2%) in the conventionally managed group.and among eyes that had zone II, stage 3 ROP with plus Nystagmus occurred in 22% of randomized infants with 239
  8. 8. Good et al TABLE 9. STRUCTURAL OUTCOME AT NINE MONTHS FOR INFANTS WITH BILATERAL HIGH-RISK PRETHRESHOLD RETINOPATHY OF PREMATURITY BY ICROP CATEGORY AND RM-ROP2 RISK* ICROP CLASSIFICATION EYES TREATED AT CONVENTIONALLY DISCORDANT PAIRSZONE STAGE PLUS DISEASE HIGH-RISK PRETHRESHOLD MANAGED EYES A† B‡I 3 Yes or No 27 (29.6) 27 (55.6) 8 1I 1 or 2 Yes 9 (22.2) 9 (22.2) 0 0I 1 or 2 No 75 (2.7) 75 (9.3) 6 1II 3 Yes 109 (7.3) 110 (10.9) 6 3II 3 No 3 (0) 3 (0) 0 0II 2 Yes 34 (20.6) 34 (20.6) 1 1RM-ROP2 risk0.15 to <0.30 109 (5.5) 109 (7.3) 4 20.30 to <0.45 74 (9.5) 74 (17.6) 7 10.45 81 (18.5) 82(30.5) 12 3ICROP, International Classification of Retinopathy of Prematurity20; RM-ROP2,7 a risk analysis program based on natural history data from theMulticenter Trial of Cryotherapy for Retinopathy of Prematurity Study.*Data are presented as number (percentage unfavorable) unless otherwise indicated. High risk was 0.15 or greater. Plus disease was defined as a degree of dilation and tortuosity of the posterior retinal blood vessels meeting or exceeding that of a standard photograph. Data include only those eyes for which structural outcome was able to be graded.†For group A, earlier treated eyes had a favorable outcome, and conventionally managed eyes had an unfavorable outcome.‡For group B, earlier treated eyes had an unfavorable outcome, and conventionally managed eyes had a favorable outcome. TABLE 10. COMPLICATIONS IN EYES TREATED FOR ROP EARLY OR AT THRESHOLD* EYES TREATED EYES TREATED AT HIGH-RISK PRETHRESHOLD† CONVENTIONALLY AT THRESHOLD†COMPLICATION (n = 361) (n = 236) Intraoperative ocular complications Conjunctival or subjunctival hematoma 30 (8.3) 16 (6.8) Conjunctival laceration, unintended 16 (4.4) 5 (2.1) Hemorrhage (retinal, preretinal, vitreous) 14 (3.9) 12 (5.1) Closure of the central retinal artery 0 (0.0) 2 (0.8) Inadvertent burn or freeze to area outside of 2 (0.6) 1 (0.4) target zone Other 2 (0.6) 0 (0)Systemic complications Apnea, bradycardia, or arrhythmia 31 (8.6) 10 (4.2) Acquired or increased cyanosis 13 (3.6) 4 (1.7)Need for reintubation within 10 days of 40 (11.1) 12 (5.1) treatment after stopping artificial ventilator Other 1 (0.3) 1 (0.4)*Data are presented as number (percentage).†Mean ± SD postmenstrual age at treatment in the early treatment group was 35.2 ± 2.3 weeks and in the threshold treatment group was 37.0 ± 2.5 weeks.bilateral high-risk ROP. compared with an average postmenstrual age at treatment Table 10 compares other ocular and systemic compli- of 37.0 weeks in conventionally managed eyes that under-cations of treatment among eyes and infants treated at went peripheral retinal ablation at threshold ROP.high-risk prethreshold versus conventionally managedhigh-risk prethreshold eyes that progressed and later DISCUSSIONunderwent treatment at threshold ROP. Ocular complica-tion rates were similar in the two groups. Systemic This study of treatment for high-risk prethreshold ROPcomplications were higher following treatment at high- showed a benefit of earlier treatment compared torisk prethreshold. High-risk prethreshold eyes random- conventional management in the primary outcome meas-ized to early treatment received peripheral retinal abla- ure of grating visual acuity at 9 months corrected age, andtion at an average postmenstrual age of 35.2 weeks, a much greater benefit for structural outcome at 6 and 9240
  9. 9. Final Results of the Early Treatment for Retinopathy of Prematurity (ETROP) Randomized Trialmonths corrected age. Whereas the rates of ophthalmo- of zone I cases to advances in neonatal care and improvedlogic complications were similar among the two treatment survival rates of the smallest premature infants. However,arms, infants in the high-risk prethreshold treatment a thorough analysis (not presented here) of the data fromgroup were more likely to experience systemic complica- the two studies (CRYO-ROP and ETROP) showed thattions of apnea, bradycardia, or reintubation following even when the effects of birth weight and gestational ageearlier treatment than with treatment at conventional are controlled, the number of zone I eyes in the ETROPthreshold, perhaps due to the earlier average postmen- cohort is still significantly higher than in the CRYO-ROPstrual age at which treatment was conducted. There was study. Perhaps other changes in the care of prematureno mortality or known permanent morbidity attributed to infants, as yet unrecognized, have given rise to an increasetreatment in either group. in zone I disease and a decrease in its severity. This report includes all the 9-month visual acuity and An alternative explanation is that examiners may nowstructural outcome examinations (98.2% of infants followed be more attentive to diagnosing zone I ROP than theyto age 9 months corrected). The beneficial effect of treat- were previously. The CRYO-ROP study showed a clearment at high-risk prethreshold ROP on structural outcome benefit of retinal ablative therapy, but the results in zoneand on visual acuity outcome is even stronger than origi- I eyes were not impressive, since most of these eyes devel-nally reported and provides further support for treatment oped unfavorable visual and structural outcomes evenof certain prethreshold eyes (type 1, see below) and careful after receiving treatment at threshold. After publication ofobservation of other prethreshold eyes (type 2).8 the CRYO-ROP results, it is possible that eyes were more In the ETROP study, a novel risk model that was carefully monitored and observed by ophthalmologists,developed based on natural history data from the CRYO- and that some eyes diagnosed as zone I today might haveROP study was used to identify infants at high risk for been categorized as zone II in the era before treatmentadverse outcomes from ROP, and only those infants were was proven effective. Additionally, some prior investiga-randomized.5-7 The model is available at tors have considered posterior zone II and zone I eyes to be in the same category.4 An assignment of posterior zoneThe model used demographic characteristics of the infant II eyes to the zone I category could have the effect ofand clinical features of the ROP to classify eyes with increasing the number of zone I eyes in this study. Theseprethreshold ROP as high-risk or low-risk. The validity of subtle factors could explain both the increased frequencythe model is demonstrated in the finding that high-risk and the improved outcome of zone I eyes in ETROPprethreshold eyes that received conventional manage- subjects compared to CRYO-ROP subjects.ment showed a much higher percentage of progression to For all groups of eyes in the ETROP study, the effectthreshold disease than those at low risk (66.4% versus of treatment at high-risk prethreshold is more15.5%, respectively) and a much higher percentage of pronounced for structural outcomes than for visual acuityunfavorable structural outcome (10.0% versus 1.3%, outcomes. A similar discrepancy between the magnituderespectively, at 6 months). Eyes with low-risk prethresh- of the difference between treatment groups for visualold ROP were managed conventionally, with treatment acuity versus structural outcomes was also observed in theadministered if conventional threshold was reached. CRYO-ROP study.1,17,18 The ETROP study chose visualOverall, study data support treatment of only selected eyes acuity as its primary outcome because vision is the mostthat develop prethreshold ROP. important measure of a treatment designed to prevent In the CRYO-ROP study, only about 9.6% of eyes in visual loss from severe ROP. Also, there was a safetythe natural history cohort with prethreshold ROP had concern that treatment at high-risk prethreshold withzone I disease, and of these eyes, 33.3% had an unfavor- laser could have some previously unrecognized deleteri-able structural outcome.21 This produced strong weighting ous effect on visual acuity. The Teller acuity card proce-on the risk factor, presence of disease in zone I, in the RM- dure was selected as the assessment tool for measurementROP2 model. In the ETROP study, 22.7% of eyes with of visual acuity at the 9-month examination because itprethreshold ROP had zone I disease, and because of the allows quantification of visual acuity in infants andstrong weighting of zone I disease in the risk model, because it had been used successfully to test infants of a94.7% of these eyes were classified as high-risk. These similar age in the CRYO-ROP study.11,17eyes represented about 40% of eyes in the ETROP As in the CRYO-ROP study,1,17,18 the finding of arandomized trial. discrepancy between the magnitude of the treatment The difference between the CRYO-ROP study and group differences in visual acuity versus structuralthe ETROP study in frequency of zone I disease and in the outcomes in the ETROP study is likely to be due, in part,more benign course of zone I disease in the ETROP study to non-ROP-related ophthalmologic and neurologic prob-is noteworthy. It is tempting to attribute the large number lems that can occur in very premature infants. Children 241
  10. 10. Good et alwith severe ROP may develop visual impairment second- treatment for those eyes that need it?” Based on the struc-ary to neural insult or other cerebral factors.22,23 In addi- tural outcome data at 6 months for a cohort of eyes withtion, nystagmus, which reduces visual acuity, was found in prethreshold ROP of all degrees of severity that wereover 20% of randomized infants with bilateral disease. conventionally managed, it is possible to determine theThese non-ROP-related factors may have resulted in number of eyes with high-risk prethreshold ROP that hadreduced acuity in conventionally managed eyes, as well as a favorable outcome without peripheral retinal ablation,in eyes treated at high-risk prethreshold ROP, thereby as illustrated in Appendix 1. Table A in Appendix 1decreasing the difference in visual acuity outcome summarizes the results of an analysis of data from thebetween the two groups of eyes. natural history cohort of prethreshold eyes in the ETROP Another likely contributor to the difference between study. The table shows that 136 (36.6%) of the 372 high-functional and structural outcomes in this study is the risk prethreshold eyes in the conventionally managedimmaturity of the visual system at 9 months post-term. group that were examined at 6 months had favorableBecause the visual acuity of a 9-month-old infant is well structural outcomes and never developed threshold ROP.below that of a normal adult, it is possible that some visual That is, these eyes met the criteria for early treatment,deficits that result from structural abnormalities will not yet, without treatment, went on to a favorable outcome atbe apparent until older ages, when acuity in normal eyes 6 months.has improved to near-adult levels. Follow-up testing using To reduce the number of eyes treated that wouldrecognition (letter) visual acuity charts at older ages would have had a favorable outcome without intervention, addi-be expected to reveal these deficits in visual acuity, as it tional strategies for selecting eyes for earlier treatmentdid in the CRYO-ROP study.1,18 were explored using this same cohort of prethreshold In evaluating the benefit of treatment at high-risk eyes. Table B in Appendix 1 indicates that eyes with zoneprethreshold, it is important to take into account possible I, stage 1 or 2 without plus disease, as well as eyes withadverse effects and trade-offs related to earlier treatment. zone II, stage 3 without plus disease, had lower rates ofThese include an increased rate of systemic complica- progressing to threshold or unfavorable outcome thantions, potential long-term risks of earlier treatment, an eyes in the other ICROP categories. And, when treated atincrease in the number of eye examinations needed to the conventional threshold, those two groups of eyes haddetect prethreshold ROP, and an increased frequency of less than 5% unfavorable structure outcomes.treatment of eyes that would otherwise have undergone The results of this analysis, along with the results inspontaneous regression of ROP. In the following para- Tables 8 and 9, led to a clinical algorithm in which treat-graphs, we discuss these issues. ment should be considered for eyes with zone I, any stage In the ETROP study, systemic complications, includ- ROP with plus disease; eyes with zone I, stage 3 ROPing apnea, bradycardia, and reintubation, occurred more without plus disease; and eyes with zone II, stage 2 or 3frequently when peripheral retinal ablative therapy was with plus disease. As shown in Table C of Appendix 1, useperformed at high-risk prethreshold than at conventional of this ICROP-based, limited selection algorithm wouldthreshold, probably because of the younger average post- have resulted in treatment of 91 eyes that showed favor-menstrual age at which the treatment of high-risk able outcomes and never reached threshold disease. Thisprethreshold eyes occurred. Ophthalmic complications is a reduction of 33% from the 136 such eyes that wouldfollowing retinal ablative therapy were comparable in eyes have been treated using the RM-ROP2 risk model astreated at high-risk prethreshold and conventionally applied in the ETROP study (Table A in Appendix 1).managed eyes, as were ophthalmic complications (other If it is assumed that conventional threshold ROPthan retinal detachment) when the entire group of continues to occur in 6% of infants weighing less thanconventionally managed eyes was compared to the group 1,251 g at birth, as in the CRYO-ROP study,24 the earlyof eyes treated at high-risk prethreshold. One potential treatment algorithm based on RM-ROP2 would result indeleterious effect of earlier treatment that was not evalu- treatment of 9% of infants. The ICROP-based, limitedated in this study is the effect of peripheral retinal ablation selection criteria described in the preceding paragraphon peripheral vision. It is possible that ablation in zone I would result in treatment of 8% of infants while retainingwill result in a greater loss in visual field extent than the advantage of early treatment for eyes at highest risk ofperipheral ablation in zone II. adverse outcomes. Another issue related to earlier treatment of ROP An alternative approach to using the ICROP-based,concerns the treatment of eyes that would have under- limited selection algorithm is to base treatment on thegone spontaneous involution without treatment. The RM-ROP2 risk model using a risk of ≥30%, instead ofquestion arises: “How many eyes must receive treatment ≥15% as the criterion for early treatment. Since theunnecessarily in order to achieve the benefit of earlier absolute risk of an unfavorable outcome is low in the risk242
  11. 11. Final Results of the Early Treatment for Retinopathy of Prematurity (ETROP) Randomized Trialrange from 0.15 to <0.30 (Tables 8 and 9) and therefore stances, peripheral retinal ablation should be consideredthe relative benefit is not as great as in the higher-risk for any eye with:categories, use of the higher-risk criterion for treatmentwould reduce treatment of eyes that would not progress to Type I ROPthreshold, while maintaining the benefit of earlier treat- • Zone I, any stage ROP with plus disease orment in higher-risk eyes. However, using such a model • Zone I, stage 3, with or without plus disease ormay be more difficult in a clinical setting than using a • Zone II, stage 2 or 3 ROP, with plus diseaserevised treatment algorithm based on the eye findings(ICROP) alone. Plus disease, in this instance, requires at least two quad- In the ETROP protocol, timely identification of high- rants (usually six or more clock hours) of dilation andrisk prethreshold ROP was important to the successful tortuosity of the posterior retinal blood vessels and, hence,application of an early treatment program; hence, infants the presence of significant disease. The algorithm doeswere followed on a weekly basis after developing zone II, not take into account all of the other known risk factorsstage 2 ROP or if they had retinal vessel immaturity with (eg, extent of stage 3, birth weight), and therefore somevessels ending in zone I, but no ROP in zone I. Infants clinical judgment is required in applying this initial step towith low-risk prethreshold disease were followed twice the management of ROP.weekly and managed conventionally unless a change in The clinical algorithm also indicates that continuedstatus caused by development of more severe ROP serial examinations, as opposed to peripheral retinal abla-resulted in advancement into the high-risk category. Thus, tion, should be considered for any eye with:a screening program aimed at identifying eyes for treat-ment prior to conventional threshold may require an Type II ROPincrease in the number of screening examinations • Zone I, stage 1 or 2 with no plus disease orconducted in the neonatal nursery. • Zone II, stage 3 with no plus disease The long-term effects of earlier treatment for ROPare as yet unknown. Because there is considerable devel- Treatment should be considered for an eye with typeopment of visual acuity that occurs between infancy and II ROP when progression to type I status or thresholdchildhood, and because it is possible to measure aspects of ROP occurs.visual function in childhood that are not assessed easily in It is important to note that even with the addition ofinfancy, the National Eye Institute has funded continued early treatment of selected eyes with prethreshold ROP,follow-up of randomized children in the ETROP study to some eyes will still progress to an unfavorable visualthe age of 6 years. At that age, recognition visual acuity and/or structural outcome. Thus, additional research iswill be measured with Early Treatment Diabetic needed to identify better methods for the prevention andRetinopathy Study charts.25 Visual field extent, contrast treatment of severe ROP.sensitivity, and ocular status will also be evaluated, andeach child’s developmental status will be assessed. This APPENDIX 1longer follow-up will give a more detailed evaluation ofthe impact of earlier treatment on the visual, ophthalmo- The results presented in this Appendix are based on datalogic, and general developmental status of study partici- from 664 infants who had one eye identified at prethresh-pants. old ROP that was not treated unless the ROP progressed to threshold. This natural history cohort was examined toClinical Implications consider alternative treatment strategies for managing The results of this study show that it is possible to identify prethreshold eyes. Included in this natural history cohortcharacteristics of ROP that predict which eyes are most were all control eyes in the asymmetric randomizedlikely to benefit from early peripheral retinal ablation. group, the conventionally managed eyes of infants withBased on study data, a clinical algorithm was developed to bilateral high-risk prethreshold disease, and one eyeidentify for early treatment eyes with prethreshold ROP selected at random from the infants with low-riskthat are at highest risk for retinal detachment and blind- prethreshold disease. Table A categorizes these eyes byness, while minimizing treatment of prethreshold eyes RM-ROP2 high-risk and low-risk classification, and bylikely to show spontaneous regression of ROP. The use of whether or not they progressed to threshold ROP forthis algorithm circumvents the need for computer-based treatment and/or had an unfavorable outcome.calculation of low risk or high risk, as was used in this Table B shows this cohort of eyes classified by ICROPstudy. categorization and indicates that nearly 100% of zone I The clinical algorithm shows that, in most circum- eyes were classified as high-risk. However, among eyes 243
  12. 12. Good et al APPENDIX TABLE A. STRUCTURAL OUTCOMES OF LOW-RISK PRETHRESHOLD EYES AND CONVENTIONALLY MANAGED HIGH-RISK PRETHRESHOLD EYES AT SIX MONTHS POST-TERM LOW-RISK (<0.15) PRETHRESHOLD (n = 292) HIGH-RISK (≥0.15) PRETHRESHOLD (n = 372)SIX-MONTH NEVER DEVELOPED DEVELOPED THRESHOLD NEVER DEVELOPED DEVELOPED THRESHOLDOUTCOME THRESHOLD ROP ROP THRESHOLD ROP ROPFavorable 245 44 136 205Unfavorable 1 2 4 27ROP, retinopathy of prematurity. APPENDIX TABLE B. PRETHRESHOLD EYES THAT REACHED HIGH-RISK PRETHRESHOLD ROP AND PRETHRESHOLD EYES THAT REACHED THRESHOLD OR UNFAVORABLE STRUCTURAL OUTCOME AT SIX MONTHS POST-TERM HIGH-RISK (≥0.15) THRESHOLD OR UFZONE ICROP STAGE PLUS N n (%) n (%)I 3 Yes 19 19 (100.0) 19 (100.0)I 3 No 14 14 (100.0) 9 (64.3)I 1 or 2 Yes 18 18 (100.0) 15 (83.3)I 1 or 2 No 101 93 (92.1) 41 (40.6)II 3 Yes 172 167 (97.1) 118 (68.6)II 3 No 274 5 (1.8) 44 (16.1)II 2 Yes 66 56 (84.8) 37 (56.1) Total 664 372 (56.0) 283 (42.6) Type I ROP 289 274 (94.8) 198 (68.5)Type II ROP (shaded above) 375 98 (26.1) 85 (22.7)ICROP, International Classification of Retinopathy of Prematurity; ROP, retinopathy of prematurity; UF, unfavorable. APPENDIX TABLE C. EFFECT OF USING ICROP-BASED CRITERIA (TYPE I AND TYPE II ROP) TO SELECT INFANTS FOR PERIPHERAL ABLATIVE TREATMENT TYPE II PRETHRESHOLD (N = 375) TYPE I PRETHRESHOLD (N = 289)SIX-MONTH NEVER DEVELOPED DEVELOPED THRESHOLD NEVER DEVELOPED DEVELOPED THRESHOLDOUTCOME THRESHOLD ROP ROP THRESHOLD ROP ROPFavorable 290 77 91 172Unfavorable 1 7 4 22ICROP, International Classification of Retinopathy of Prematurity; ROP, retinopathy of prematurity.with zone I, stage 1 or 2 ROP with no plus disease, a much classification are shown in Table C.lower percentage progressed to threshold and/or an unfa-vorable outcome than occurred in all other categories of APPENDIX 2zone I disease. Among zone II eyes, those with zone II,stage 3 with no plus disease did much better than eyes in ETROP STUDY INVESTIGATORSother categories. These data led to a proposed grouping ofthe eyes by the ICROP classification into type I and type Writing Committee: Chair: William V. Good, MD;II ROP. The outcome results achieved by dividing the Robert J. Hardy, PhD; Velma Dobson, PhD; Earl A.cohort into type I and type II ROP based on the ICROP Palmer, MD; Dale L. Phelps, MD; Michelle Quintos, BA;244
  13. 13. Final Results of the Early Treatment for Retinopathy of Prematurity (ETROP) Randomized TrialBetty Tung, MS. Ira H. Gewolb, MD. Co-principal Investigator: Kelly A.Stanford Center (Palo Alto, California). Lucille Hutcheson, MD. Study Center Coordinators: LoniPackard Children’s Hospital, Stanford University. Co- Huynh, COA; Rani Kalsi, BA, COA; Xiaonong Liu; L.principal Investigators: Ashima Madan, MD; Michael Jennifer Smell, RN. Coinvestigators: Susan J. Dulkerian,Gaynon, MD. Study Center Coordinators: M. Bethany MD; Michael J. Elman, MD; Eric Jones, MD; Mark W.Ball, BS; Patricia N. Hartsell, RN, BA; Dottie Inguillo, Preslan, MD; Scott M. Steidl, MD, DMA.RN. Coinvestigators: Deborah Alcorn, MD; William V. Baltimore R Center, Baltimore, Maryland). JohnsGood, MD; Donna Ornitz, MD; David Stevenson, MD. Hopkins Hospital, Johns Hopkins Bayview MedicalSan Francisco Center (San Francisco, California). Center, Howard County General Hospital, GreaterCalifornia Pacific Medical Center, Oakland Children’s Baltimore Medical Center, St Joseph Medical Center.Hospital, University of California, San Francisco, Medical Principal Investigator: Michael X. Repka, MD. StudyCenter. Principal Investigator: William V. Good, MD. Center Coordinators: Jennifer A. Shepard, NNP; PamelaStudy Center Coordinators: Monica Hubbard, MS, PNP; Donahue, PhD. Coinvestigators: Susan W. Aucott, MD;Jason Lee, MD. Coinvestigators: Daniel Brinton, MD; Tuvia Blechman, MD; Mary Louise Collins, MD;Susan Day, MD; David Durand, MD; Douglas Fredrick, Maureen M. Gilmore, MD; James T. Handa, MD; AnanthMD; Roderic H. Phibbs, MD; Daniel Schwartz, MD; Vijay Mudgil, MD; Quan Dong Nguyen, MD; Cameron F.Terri Slagle, MD; Gordon Smith, MD. Parsa, MD; Dante Pieramici, MD; David Plotsky, MD;Chicago Center (Chicago, Illinois). University of Jeffrey J. Pomerance, MD.Illinois at Chicago Hospital and Medical Center. Principal Boston Center (Boston, Massachusetts). NewInvestigator: Michael Shapiro, MD. Study Center England Medical Center, Children’s Hospital, BrighamCoordinators: Yesenia Garcia; Maria Genio; Jeffrey and Women’s Hospital, Beth Israel Deaconess MedicalParker; Bernadine Rupar. Coinvestigators: Herbert Center, Lowell General Hospital, Lawrence GeneralBecker, MD; Rama Bhat, MD; Jeffrey N. Bloom, MD; Hospital, Winchester Hospital, Newton-WellesleyJessica V. Corsino, MD; Lawrence Kaufman, MD; Wico Hospital, South Shore Hospital, Melrose-WakefieldWaikwan Lai, MD; Jose Pulido, MD, MS; Tonse N.K. Hospital, Beverly Hospital. Principal Investigator:Raju, MD; Arvid K. Shukla, MD; Benjamin Ticho, MD; Cynthia H. Cole, MD, MPH. Co-principal Investigator:Dharmapuri Vidyasagar, MD. Deborah Vanderveen, MD; Study Center Coordinators:Indianapolis Center (Indianapolis, Indiana). Lacy Berman; Christy Faherty, RN, BSN; Caitlin Hurley,(Indiana University School of Medicine) James Whitcomb BS; Terry Mansfield, RN; Brenda McKinnon, RNC;Riley Hospital for Children, Indiana University Hospital, Marianne Moore, RN. Coinvestigators: Caroline Baumal,Wishard Memorial Hospital, Methodist Hospital, MD, FRCSC; Amita Bhatt, MD; Mark Dacey, MD; JayCommunity Hospitals of Indianapolis. Principal Duker, MD; Janine Eagle, MD; Anthony Fraioli, MD;Investigator: James Lemons, MD. Co-principal Paul Greenberg, MD; Mark Hughes, MD; Robert Lacy,Investigator: Daniel Neely, MD. Study Center MD; O’ine McCabe, MD; Robert Peterson, MD; EliasCoordinators: Dee Dee Appel, RN; Elizabeth A. Hynes, Reichel, MD; Adam Rogers, MD; William Stinson, MD;RN; Leslie Wright, RN. Coinvestigators: David Plager, Mitchell Strominger, MD.MD; Naval Sondhi, MD; Derek Sprunger, MD. Detroit Center (Detroit, Michigan). WilliamLouisville Center (Louisville, Kentucky). Kosair Beaumont Hospital, Children’s Hospital of Michigan, StChildren’s Hospital, University of Louisville Hospital. John’s Hospital Detroit. Principal Investigator: JohnPrincipal Investigator: Charles C. Barr, MD. Study Center Baker, MD. Study Center Coordinators: Kristi Cumming,Coordinator: Greg K. Whittington, PsyS. Coinvestigators: MSN; Michelle Kulak RN; Pat Manatrey, RN.Marianne Cowley, MD; Craig H. Douglas, MD; Peggy H. Coinvestigators: Daniel Batton, MD; Mary Bedard, MD;Fishman, MD; Tonya Robinson, MD; Paul J. Rychwalski, Antonio Capone, MD; Renato Casabar, MD; EdwardMD. O’Malley, MD; Rajesh Rao, MD; John Roarty, MD;New Orleans Center (New Orleans, Louisiana). Michael Trese, MD; George Williams, MD.Tulane University Medical Center, Medical Center of Minneapolis Center (Minneapolis, Minnesota).Louisiana at New Orleans. Principal Investigator: Robert Fairview University Medical Center, Children’s HealthA. Gordon, MD. Study Center Coordinator: Deborah S. Care, Minneapolis, Hennepin County Medical Center.Neff, LPN. Coinvestigators: Douglas B. Babel, OD, MD; Principal Investigator: Stephen P. Christiansen, MD.James G. Diamond, MD; William L. Gill, MD. Study Center Coordinators: Sally Cook, BA; AnnBaltimore G Center (Baltimore, Maryland). Holleschau, BA; Molly Maxwell, RN; Marla Mills, RN,University of Maryland Medical Systems, Mercy Medical MSN; Carol Miller, RN; Kristin Rebertus, RN, NNP;Center, Franklin Square Hospital. Principal Investigator: Nancy Trower, RN, NNP. Coinvestigators: Steven 245
  14. 14. Good et alBennett, MD; David Brasel, MD; Robert Couser, MD; MD; Leon-Paul Noel, MD; Robert Olsen, MD; SuzanneSundeep Dev, MD; Allison Jensen, MD; Richard Lussky, Pesce, MD; Steven Rose, MD; Bryan Rutledge, MD;MD; George Miller, MD; Robert Mittra, MD; Timothy Richard Simon, MD; Sam Spalding, MD; Donald Tingley,Olsen, MD; Robert Ramsey, MD; William Rosen, MD; MD; Paul Torrisi, MD; Robert Vanderlinde, MD.Edwin Ryan, MD; Shelley Springer, MD; Eric Steuer, Columbus Center (Columbus, Ohio). ColumbusMD; C. Gail Summers, MD; David Williams, MD. Children’s Hospital, Ohio State University Hospital,St Louis Center (St Louis, Missouri). Cardinal Mount Carmel Medical Center, Grant Medical Center,Glennon Children’s Hospital, St Mary’s Health Center. Riverside Methodist Hospital, Mount Carmel EastPrincipal Investigator: Bradley V. Davitt, MD. Study Hospital, St Ann’s Hospital. Principal Investigator: GaryCenter Coordinators: Julie Beuer, RN; Linda Breuer, L. Rogers, MD. Co-principal Investigator: Don Bremer,LPN. Coinvestigators: Oscar Cruz, MD; Stephen Feman, MD. Study Center Coordinators: Rae Fellows, MEd;MD; William Keenan, MD; Greg Mantych, MD. Sharon Klamfoth, LPN; Brenda Mann, RNC;North Carolina Center (Durham and Chapel Hill, Coinvestigators: Leandro Cordero, MD; Richard Hertle,North Carolina). Duke University Medical Center, MD; Alan Letson, MD; Richard McClead, MD; MaryUniversity of North Carolina Hospital. Principal Lou McGregor, MD; Patrick Wall, MD.Investigator: Sharon Freedman, MD. Co-principal Oklahoma City Center (Oklahoma City, Oklahoma).Investigator: David Wallace, MD. Study Center Children’s Hospital of Oklahoma. Principal Investigator:Coordinators: Eileen Camp, RN; Sharon Clark, RN; Lori R. Michael Siatkowski, MD. Study Center Coordinators:Hutchins, RN; Lora Lake, RN. Coinvestigators: Edward Karen E. Corff, MS, ARNP; Melissa Fuhr, RN.Buckley, MD; Laura Enyedi, MD; Ricki Goldstein, MD; Coinvestigators: Reagan H. Bradford, MD; Robert E.Maurice Landers III, MD; Diane Marshall, MD; Travis Leonard, MD; Mark H. Scott, MD.Meredith, MD; Kean Oh, MD; Joan Roberts, MD. Portland Center (Portland, Oregon). DoernbecherBuffalo Center (Buffalo, New York). Women’s and Children’s Hospital at Oregon Health and ScienceChildren’s Hospital of Buffalo, Sisters of Charity Hospital. University, Legacy Emanuel Children’s Hospital,Principal Investigator: James D. Reynolds, MD. Study Providence St Vincent’s Hospital. Principal Investigator:Center Coordinators: Dawn C. Gordon, RNC; Barbara David T. Wheeler, MD. Study Center Coordinators:Kuppel, RN, BSN. Coinvestigators: George P. Albert, Karen Davis, RN; Nancy Dolphin, RN; Sharon Dunham,MD; Steven Awner, MD; Rita Ryan, MD. RN. Coinvestigators: Aazy Aaby, MD; Shawn Goodman,Long Island/Westchester Center (New York). Stony MD; Andreas Lauer, MD; Valerie Newman, MD; Earl A.Brook University Hospital, Westchester Medical Center. Palmer, MD; De-Ann Pillers, MD, PhD; JosephPrincipal Investigator: Pamela Ann Weber, MD. Study Robertson, MD; Ann Stout, MD; Tim Stout, MD; AndreaCenter Coordinators: Adriann Combs, RNC; Natalie Tongue, MD.Dweck, RN. Coinvestigators: Howard Charles, MD; Tina Philadelphia Center (Philadelphia, Pennsylvania).Chou, MD; Joseph DeCristofaro, MD; Corina Gerontis, The Children’s Hospital of Philadelphia, The Hospital ofMD; Marc Horowitz, MD; Richard Koty, MD; Edmund the University of Pennsylvania, Pennsylvania Hospital.LaGamma, MD; Maury Marmor, MD. Principal Investigator: Graham E. Quinn, MD, MSCE.New York Center (New York, New York). New York Study Center Coordinators: Jamie G. Koh, RN, MSN,Presbyterian Hospital, Columbia Campus, New York CCRC; Marianne E. Letterio, RN, BSN; MollyPresbyterian Hospital, Cornell Campus. Principal McDaniel, BA. Coinvestigators: Soraya Abbasi, MD; JaneInvestigator: John Flynn, MD. Co-principal Investigator: C. Edmond, MD; Brian J. Forbes, MD, PhD; Albert M.Thomas Lee, MD. Study Center Coordinators: Osode Maguire, MD; Monte D. Mills, MD; Eric A. Pierce, MD,Coki, RNC, BSN. Coinvestigators: Michael Chiang, MD; PhD; Terri L. Young, MD.Steven Kane, MD; Alfred Krauss, MD; Robert Lopez, Pittsburgh Center (Pittsburgh, Pennsylvania).MD; Richard Polin, MD. Magee-Women’s Hospital. Principal Investigator:Rochester/Syracuse Center (New York). University of Kenneth Cheng, MD. Study Center Coordinator: JudithRochester Medical Center, Crouse-Irving Memorial Jones, RNC, BSN. Coinvestigators: Robert Bergren, MD;Hospital. Principal Investigator: Dale L. Phelps, MD. Co- Beverly Brozanski, MD; Bernard Doft, MD; Mitchellprincipal Investigators: Steven J. Gross, MD; David Fineman, MD; Louis Lobes, MD; Karl Olsen, MD.Hakanson, MD. Study Center Coordinators: Marcia Charleston Center (Charleston, South Carolina).Dodge, RN; Cassandra Horihan, MS; Pamela Parker, BA; Medical University of South Carolina. PrincipalJane Phillips. Coinvestigators: Dennis Asselin, MD; Shi- Investigator: Richard A. Saunders, MD. Study CenterHwa W. Chang, MD; Ernest Guillet, MD; Robert Coordinator: Lisa Langdale, RN. Coinvestigators: AmyHampton, MD; Gary Markowitz, MD; Walter Merriam, Hutchinson, MD; M. Millicent Peterseim, MD; Dilip246
  15. 15. Final Results of the Early Treatment for Retinopathy of Prematurity (ETROP) Randomized TrialPurohit, MD. Members: William V. Good, MD; Robert J. Hardy, PhD;Houston Center (Houston, Texas). Baylor College of Maryann Redford, DDS, MPH.Medicine, Texas Children’s Hospital, Texas Woman’s Executive Committee, Permanent Members: Chair:Hospital, Ben Taub General Hospital. Principal William V. Good, MD; Robert J. Hardy, PhD; VelmaInvestigator: David K. Coats, MD. Study Center Dobson, PhD; Earl A. Palmer, MD; Dale L. Phelps, MD;Coordinators: Laura Gonzalez; Nataliya Kazymyrko, MD; Ex-officio member: Maryann Redford, DDS, MPH.Alma Sanchez, COT; Michele Steward, COT. Executive Committee, Elected Members: W.A.J. vanCoinvestigators: Kathryn Brady-McCreery, MD; Joseph Heuven, MD (2000-2001); Charles Barr, MD (2001-Garcia-Prats, MD; Eric Holz, MD; Scott Jarriel, MD; 2002); Michael Gaynon, MD (2002-2003); MichaelKaren Johnson, MD; George Mandy, MD; Evelyn A. Shapiro, MD (2003-2004); Rae Fellows, MEd (2000-Paysee, MD; A. Melinda Rainey, MD; Kimberly G. Yen, 2001); Judith Jones, RNC, BSN (2001-2002); KristiMD. Cumming, MSN (2002-2003); Deborah S. Neff, LPNSan Antonio Center (San Antonio, Texas). University (2003-2004).Hospital, Christus Santa Rosa Children’s Hospital. Editorial Committee: Chair: William V. Good, MD;Principal Investigator: W.A.J. van Heuven. Co-principal Robert J. Hardy, PhD; Velma Dobson, PhD; Earl A.Investigator: Alice K. Gong, MD. Study Center Palmer, MD; Dale L. Phelps, MD; Michelle Quintos, BA;Coordinator: Melanie H. Drummond, RN. Betty Tung, MS.Coinvestigators: Timothy Paul Cleland, MD; James C.MacDonald, MD; Lina M. Marouf, MD; Juan Elian REFERENCESRubio, MD.Salt Lake City Center (Salt Lake City, Utah). 1. Cryotherapy for Retinopathy of Prematurity CooperativeUniversity of Utah Health Science Center, Primary Group. Multicenter trial of cryotherapy for retinopathy ofChildren’s Medical Center. Principal Investigator: Robert prematurity: ophthalmological outcomes at 10 years. ArchHoffman, MD. Study Center Coordinator: Susan Ophthalmol 2001;119:1110-1118.Bracken, RN. Coinvestigators: Paul Bernstein, MD; 2. Cryotherapy for Retinopathy of Prematurity CooperativeDavid Dries, MD; Jerald King, MD; Richard Olson, MD; Group. Multicenter trial of cryotherapy for retinopathy ofMichael Teske, MD; Kimberly Yen, MD. prematurity: preliminary results. Arch OphthalmolNational Eye Institute, Bethesda, Maryland. 1988;106:471-479.Program Officer: Maryann Redford, DDS, MPH (June, 3. Rescan Z, Vamos R, Salacz G. Laser treatment of zone I2001-Present); Richard L. Mowery, PhD (October, 2000- prethreshold and stage 3 threshold retinopathy of prematu-May, 2001); Donald F. Everett, MA (September, 1999- rity. J Ophthalmol Strabismus 2003;40:204-207.September, 2000). 4. Vander JF, Handa J, McNamara JA, et al. Early treatmentStudy Headquarters: Smith-Kettlewell Eye Research of posterior retinopathy of prematurity: a controlled trial.Institute, San Francisco, California. Principal Ophthalmology 1997;104:1731-1735; discussion 1735-1736.Investigator: William V. Good, MD. Project Coordinator: 5. Good WV, Hardy RJ, for the ETROP Multicenter StudyMichelle Quintos, BA. Group. The multicenter study of early treatment forCoordinating Center: School of Public Health, retinopathy of prematurity (ETROP) [guest editorial].Coordinating Center for Clinical Trials, University of Ophthalmology 2001;108:1013-1014. 6. Early Treatment for Retinopathy of PrematurityTexas Health Science Center, Houston, Texas. Principal Cooperative Group. Multicenter trial of Early TreatmentInvestigator: Robert J. Hardy, PhD. Project Manager: for Retinopathy of Prematurity: study design. Contr ClinBetty Tung, MS. Coordinating Center Staff: Charles Trials 2004;25:311-325.Cooper, MS; Gordon Tsai, MS; Meng-Fen Wu, MS; 7. Hardy RJ, Palmer EA, Dobson V, et al. Risk analysis ofCharles Minard, MS; Krystal Rather, BS. prethreshold ROP. Arch Ophthalmol 2003;121:1699-1701.Vision Testing Center: University of Arizona, School of 8. Early Treatment for Retinopathy of PrematurityMedicine, Tucson, Arizona. Principal Investigator: Velma Cooperative Group. Revised indications for the treatmentDobson, PhD. Coinvestigator: Graham E. Quinn, MD. of retinopathy of prematurity: results of the early treatmentVision Testers: Kathleen M. Mohan, MA; Meigan B. for retinopathy of prematurity randomized trial. Arch Ophthalmol 2003;121:1684-1696.Baldwin, BA. Vision Testing Center Coordinator: Suzanne 9. McDonald MA, Dobson V, Series SL, et al. The acuity cardM. Delaney, PhD. procedure: a rapid test of infant acuity. Invest OphthalmolData and Safety Monitoring Committee: Chair:John Vis Sci 1985;26:1158-1162.Connett, PhD. Members: Edward F. Donovan, MD; 10. Teller DY, McDonald MA, Preston K, et al. Assessment ofArgye Hillis, PhD; Jonathan M. Holmes, MD; Joseph M. visual acuity in infants and children: the acuity card proce-Miller, MD; Carol R. Taylor, RN, CSFN, PhD. Ex-officio dure. Dev Med Child Neurol 1986;28:779-789. 247
  16. 16. Good et al11. Dobson V, Quinn GE, Biglan AW, et al. Acuity card assessment Good’s paper detailing the outcome of the Early of visual function in the cryotherapy for retinopathy of prema- Treatment for ROP (ETROP) Randomized Clinical Trial, turity trial. Invest Ophthalmol Vis Sci 1990;31:1702-1708. the latest in a series of successful trials to ask and answer12. Trueb L, Evans J, Hammel A, et al. Assessing visual acuity important questions about the disease ROP. Surely this of visually impaired children using the Teller acuity card must place such trials, CRYO-ROP, LIGHT ROP, STOP procedure. Am Orthopt J 1992;42:149-154.13. Salomão SR, Ventura DF. Large sample population age ROP, and ETROP among the most successful interven- norms for visual acuities obtained with Vistech-Teller tion trials supported by taxpayer monies through the Acuity Cards. Invest Ophthalmol Vis Sci 1995; 36:657-670. medium of the National Eye Institute.14. Mayer DL, Beiser AS, Warner AF, et al. Monocular acuity When discussing such a massive trial in a few short norms for the Teller Acuity Cards between ages one month minutes it seems to me best to try to focus on a few ques- and four years. Invest Ophthalmol Vis Sci 1995;36:671-685. tions that get to the heart of the effort involving 2615. Dobson V, Quinn GE, Siatkowski RM, et al, for the centers, 240 investigators and 823 infants studied at a cost Cryotherapy for Retinopathy of Prematurity Cooperative to the taxpayer of 13 millions of dollars. Group. Agreement between grating acuity at age 1 year and Is the disease studied important? The answer is yes. Snellen acuity at age 5.5 years in the preterm child. Invest Ophthalmol Vis Sci 1999;40:496-503. ROP once thought banished as an ophthalmic and pedi-16. Frankenburg WK, Dodds JB. The Denver developmental atric curiosity is back with a vengeance. It is among the screening test. J Pediatr 1967;71:181-191. three top causes of blindness in infancy and is almost17. Cryotherapy for Retinopathy of Prematurity Cooperative world-wide (first and second world) in its distribution. It Group. Multicenter trial of cryotherapy for retinopathy of is very much a paradox of medical progress: As modern prematurity. One-year outcome—structure and function. neonatology has advanced in its sophistication, the result Arch Ophthalmol 1990;108:1408-1416. has been the salvage of many infants well under a kilo-18. Cryotherapy for Retinopathy of Prematurity Cooperative gram in birth weight and 30 weeks gestational age. This Group. Multicenter trial of cryotherapy for retinopathy of has provided an almost inexhaustible pool of infants prematurity. Snellen visual acuity and structural outcome at 5-1/2 years after randomization. Arch Ophthalmol vulnerable to the most severe forms of ROP in numbers 1996;114:417-424. never seen before.19. Hardy RJ, Davis BR. The design, analysis, and monitoring Is the question asked an important one? Once again of an ophthalmological clinical trial. ASA Proceedings of the the answer is yes, emphatically yes. Should the threshold Biopharmaceutical Section 1989;248-253. for treatment be lowered to include infants previously20. Committee for the Classification of Retinopathy of classed as not yet at threshold but at pre- threshold? Prematurity. An international classification of retinopathy Should they be treated at this new level of severity of of prematurity. Arch Ophthalmol 1984;102:1130-1134. disease? The downside risk here would be subjecting21. Cryotherapy for Retinopathy of Prematurity Cooperative some (an unknown number) of these infants to a treat- Group. The natural ocular outcome of premature birth and retinopathy: status at 1 year. Arch Ophthalmol ment that might not indeed be necessary. 1994;112:903-912. Is the study design adequate to provide an unequivo-22. Msall ME, Phelps DL, DiGaudio KM, et al, on behalf of cal answer? The Randomized Clinical Trial is the gold the Cryotherapy for Retinopathy of Prematurity standard in assessing the benefit of intervention or treat- Cooperative Group. Severity of neonatal retinopathy of ment outcomes in medicine today. This is a lesson learned prematurity is predictive of neurodevelopmental functional early and well by the cadre of investigators who have outcome at age 5.5 years. Pediatrics 2000;106:998-1005. become addicted to the disease (ROP) and its treatment23. Jacobson L, Fernell E, Broberger U, et al. Children with over the course of the last two decades. blindness due to retinopathy of prematurity: a population- Are the results significant? Both clinically and statis- based study. Perinatal data, neurological and ophthalmolog- ical outcome. Dev Med Child Neurol 1998;40:155-159. tically they are. Most importantly what gives them that24. Palmer EA, Flynn JT, Hardy RJ, et al, for the Cryotherapy significance clinically? This study includes in its sample for Retinopathy of Prematurity Cooperative Group. 188 eyes (23%) with Zone 1 disease in contrast to CRYO- Incidence and early course of retinopathy of prematurity. ROP that had 33 eyes (12%) eyes with Zone 1 disease. Ophthalmology 1991;98:1628-1640. The results of treatment of Zone 1 eyes in the CRYO25. Ferris FL III, Kassoff A, Bresnick GH, et al. New visual study were dismal; 87% had an unfavorable response to acuity charts for clinical research. Am J Ophthalmol treatment, little better than no therapy at all, where 93% 1982;94:91-96. were unfavorable. In the current trial, Zone 1 high risk eyes had seven more discordant pairs where the earlyDISCUSSSION treatment produces a favorable response in that eye compared to an unfavorable outcome in its fellow eyeDR JOHN T. FLYNN. It is a privilege to discuss Dr Bill treated at threshold versus only one pair where the248